agentic-engineering-skills
AGENTS.md3 starsChanged 16 days agoMITView on GitHub
this is a public snapshot of a working engineering setup. share the skills and workflow so readers can adopt the parts that fit their own projects never push without explicit user permission use these commit messages for skill changes, replacing /skill with the skill's name: for other changes, use similarly short, plain conventional commits with an appropriate type, e.g. fix: correct broken link, docs: update readme or chore: update gitignore write commit messages in lowercase. describe the change directly and…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository15 other files it gives its agents.
- agent-browser-diffSkill
- ai-code-review-catalogSkill
- claude-parallel-writersSkill
- cross-harness-sub-agentSkill
- helper-sub-agentsSkill
- llm-evaluationSkill
- loopSkill
- orchestrate-multi-pr-workSkill
- planSkill
- propose-refactorSkill
- prSkill
- record-videoSkill
- triaging-feedbackSkill
- upgrade-dependenciesSkill
- writing-skillsSkill
The file
We split this AGENTS.md into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
when updating skills, inspect both repositories' git histories to see what edits have been made to the skill from the source repository, if …
- Distinctive: line 9
keep complete skill folders at the repository root, including supporting scripts, prompts and configuration
- Distinctive: line 12
check every added or updated file for confidential information and remove it before publication
- Distinctive: line 16
keep it short and list every included skill in the table
- Distinctive: line 17
when adding a skill, write 5 distinct versions of its readme entry and apply the best one
- Distinctive: line 19
when updating a skill, suggest readme edits if needed
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
purpose
this is a public snapshot of a working engineering setup. share the skills and workflow so readers can adopt the parts that fit their own projects
maintaining skills
.skill-source-repositorycontains the path or url of the repository to copy skills from; use it unless the user asks for a different repository- when updating skills, inspect both repositories' git histories to see what edits have been made to the skill from the source repository, if any
- keep complete skill folders at the repository root, including supporting scripts, prompts and configuration
- copy source skills unchanged unless edits are needed for confidentiality or explicitly requested
- project-specific commands, conventions and references to skills not included here are intentional. keep them rather than turning the collection into a self-contained framework
- check every added or updated file for confidential information and remove it before publication
readme
- keep it short and list every included skill in the table
- when adding a skill, write 5 distinct versions of its readme entry and apply the best one
- show all 5 versions as a numbered list in the final response and clearly identify which version was applied
- when updating a skill, suggest readme edits if needed
- link upstream skills without descriptions
- retain the caveat about references to commands and skills not included here
commits
never push without explicit user permission
use these commit messages for skill changes, replacing /skill with the skill's name:
feat: update /skillfeat: add /skillfeat: delete /skill
for other changes, use similarly short, plain conventional commits with an appropriate type, e.g. fix: correct broken link, docs: update readme or chore: update gitignore
write commit messages in lowercase. describe the change directly and skip hype or filler
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.